Once recruited, the patients’ paper medical files were used to collect information on civil status, weight, height, telephone number, lifestyle, risk factors and clinical, biological, radiological and histological data. A monthly, quarterly and bi-annual check of the transaminases (alanine aminotransferase (ALT)) was carried out. Tests for HBsAg, HBeAg and HBV-DNA were performed every six months or every year. HBsAg was detected using the rapid diagnostic test (the Determine TM Abbott) or the Elisa method (Vidas®). The quantification of HBsAg was carried out by the HBs Ag II quant II Cobas method: 0.05 IU/mL to 52,000 IU/mL (log10: 2.11) and the quantification of HBV-DNA by a real-time PCR (Roche Cobas Taq Man, sensitivity threshold 20 IU/mL). An abdominal ultrasound and/or an abdominal CT scan were performed annually in the absence of cirrhosis and every six months otherwise. Liver activity and fibrosis were assessed by a liver biopsy or a liver fibrosis blood test using a FibroMeter®.

Antiviral treatment was indicated in patients with activity and/or fibrosis ≥ 2. When ALT concentration was twice the superior margin of the range in a patient with detectable viral load, antiviral treatment was also initiated regardless of the activity or the fibrosis. A family history of primary liver cancer and the ability to adhere to the treatment were also considered before the treatment initiation. Using a questionnaire and a visual analogue scale (VAS) from 0 to 10 (0: the patient stopped any medication and 10: the patient takes regularly its drugs at the right time), the adherence to the treatment was assessed over the last three months. Three categories were defined with regard to the adherence: The high adherence group (no missed dose, VAS score of 10), moderate adherence group (overall good adherence over the last three months, VAS score from 8.1 to 9.9), and low or non-adherence group (seldom drug intake or treatment completely stopped over the last three months, VAS score ≤ 8). The statistical analysis was carried out using Epi Info software.

­ HBV chronic carrier: a subject whose HBsAg test is positive for more than 6 months and the anti-hepatitis B core antibodies (anti-HbcAb; total IgG) test is positive.

­ Chronic active hepatitis: a chronic hepatitis B carrier whose ALT quantification is twice above the superior threshold of the normal range of values, continuously or in a fluctuating manner and the HBV-DNA is detectable.

­ Chronic inactive carrier: a chronic hepatitis B carrier whose ALT quantification is still within the normal range and the HBV-DNA is less than 2000 IU/ml.

­ Duration of HBsAg carriage: this is the time interval between the date of HbsAg detection and the end of our study.

­ Primary non-response: was defined as a <1-log drop after 3 months of treatment as per EASL’s (European Association for the Study of the Liver) definition [6].

­ Complete virological response: decrease of serum HBV DNA to PCR-undetectable levels after 1 year of treatment.

­ Resistance: It corresponds to the rise in serum HBV DNA levels of at least 1 − log₁₀ IU/mL compared with the lowest value during therapy (nadir value), in 2 consecutive samples 1 month apart, in patients who have previously responded and have a good treatment compliance [7].

­ Primary resistance: when viral DNA remains unchanged after three months of treatment.

­ Immunotolerance: a patient whose HBV-DNA > 20,000 IU/ml with its ALT still within the normal range.

­ HBs seroconversion: it is the loss of the HBsAg with or without the detection of the Anti-HBsAb and the Anti-HBcAb is negative.

­ HBe seroconversion: it is the loss of the HBeAg with or without the detection of the anti-HBeAb.

We categorized five biochemical and virological profiles:

­ First profile: HBV-DNA < 2000 IU/ml and ALT still within the normal range (inactive carriage);

­ Second profile: a high HBV-DNA > 20,000 IU/ml and ALT consistently above the normal range;

­ Third profile: a fluctuating viral replication and ALT values without any normalization window;

­ Fourth profile: a succession of periods of elevated ALT followed by spontaneous normalization and a fluctuating viral replication;

­ Fifth profile: the immunotolerance phase with an HBV-DNA > 20,000 IU/ml and ALT consistently within the normal range.

The sample included 148 participants for a sex ratio of 3; 69% of the participants were ≤34 years old. There were 66 (44.6%) students, and 63 (29%) workers in the formal public or private sectors. Singles or widows represented 93 (62.8%) participants. Blood donation 66 (44.6%) and routine check-up 43 (29%) were the most frequent circumstances for diagnosing HBV infection.

Abdominal ultrasound was normal in 126 (85.1%) patients. A fatty or dysmorphic liver was observed in 9 (6.1%) and 7 (4.7%) cases, respectively.

The hepatic fibrosis blood test showed an activity A ≥ 2 in 25 (83.3%) patients and fibrosis F ≥ 2 in 23 (76.7%) patients. The liver biopsy was performed in 8 (5.4%) patients; 2 patients showed no inflammatory activity (A0); 5 had a minimal activity (A1). A portal fibrosis without septa (F1) was found in three patients and three others had portal fibrosis with some septa (F2).

Sixty-three patients including 49 (77.8%) men were on treatment. The remaining participants were not taking any drug either because antiviral drugs were not yet indicated for their disease status, or because they were not able to complete the biological investigations to make the decision to start a treatment. The mean age for those under treatment was 33.8 years (range: 21 to 62). The most represented occupations were pupils or students 45 (30.2%), civil servants 42 (28.6%) and private employees 33 (22.2%). Fibrometer testing was performed in 27 patients and 14 (81.5%) had inflammatory activity greater than one.

Tenofovir was administered to 50 (79.4%) patients and lamivudine to 9 (14.3%) patients (Figure 1).

The duration of the treatments with tenofovir ranged between 1 and 9 years. Normalization of ALT was observed in 42 (84%) patients while HBV-DNA became undetectable in 24/33 patients. HBeAg negativation was observed in 16 (64%) patients after seven years of treatment.

Of the nine patients on Lamivudine, two had a complete virologic response and six had a normalization of their ALT (Table 1). Two and 9 patients lost HBeAg after 7 and 9 years of treatment, respectively. No case of HBsAg seroconversion has been observed under lamivudine treatment.

Overall 40 (63%) patients were highly adherent, 17 (27%) were in the low adherence group and 6 (10%) were moderately adherent (Figure 2). According to the occupation, religious people and housewives were the most adherent (Figure 3), followed by the private sector employees, the civil servants and the students/pupils and tradespeoples at the same proportion of 50%. The highly adherent group among farmers represented 33%. The 15 - 24 age group was the least adherent (20% of the group was adherent) while the group of the 45 - 54 years old was highly (100%) adherent. The other age groups (25 - 34; 35 - 44 and 55 - 64) had an adherence level of over 60%. After three years of treatment, 72.2% of the patients were adherent against 76.3% at 5 years; 11% and 13.2% were moderately adherent at respectively 3 and 5 years, and 16.8% and 10.5% were not at all adherent still at the same time period. Among highly or moderately adherent patients, ALT were normal and viral DNA undetectable in 76.5% and 33.3% of cases, respectively. In the low adherent group, all patients had normal or abnormal transaminases, but detectable HBV DNA.

Lamivudine was initiated in 21 patients, 9 of whom were still under the molecule at the time of our study. Ten had developed resistance including primary resistance in three patients. In two patients, lamivudine was switched to tenofovir after 4 years of treatment to prevent the development of resistance. The resistance in one year was 14.3% and 47.6% in 5 years of treatment, meaning a mean increase of 12% per year. No resistance has been observed with tenofovir. Progressively, 4/41 patients with active VHB had become stable (inactive carriage with normal ALTs and HBV-DNA < 2000 IU/mL) over a follow-up period of 1 to 6

years. HBeAg seroconversion was observed in two inactive carriers. HBeAg negativation was found in 9 patients (22% of active carriers). Thirty-seven (90.2%) patients with chronic active hepatitis were observed among the untreated population.

With a sex ratio = 3, our results were comparable to those of other studies on the topic [8] [9] and clearly indicated that VHB became chronic more often in men than in women.

The mean age was 33 years and 85.8% of the participants were younger than 40. Bougouma and Khelifa found a mean age of 37.5 and 35 years, respectively among patients with positive HBeAg. Other studies found a mean age as low as 33 years [8] [10]. In the literature [11], patients with HBeAg are generally < 40 years.

Pupils and students represented 45% of our sample, followed by civil servants and private sector employees (Figure 3). These groups used to be the target population of blood collection campaigns and/or the yearly health check-up as part of occupational medical visits. The education level of these groups was undoubtedly also a contributing factor to the high rate of representativeness

A family history of HBsAg carriage, cirrhosis or primary liver cancer was found in 12.8% of cases. The weakness of HBV screening in the general population may be the explanation here.

The virological response to tenofovir was complete in 72.7% of cases. This rate is comparable to the virological response reported in the literature, which fluctuates between 72% and 76% [12] [13]. Furthermore, the ALT normalized in 84% of our patients taking tenofovir over a period of 1 to 6 years (Table 1) versus 69% in the literature [12] for a follow-up of one year.

According to the defined categories of therapeutic adherence, 40 (63%) patients were highly adherent; 6 (10%) patients were moderately adherent and 17 (27%) patients were in the low-adherence group (Figure 2). In other countries, the high adherence level was similar. Besides, 32% and 7% of cases were in the moderate or low-adherence group [14] [15]. The high proportion of low-adherent patients in our survey could be explained by the skipped doses because of the lack of habit at the beginning of the treatment and the consecutive forgetfulness, a lack of money to supply drugs [16], the drugs shortage at the level of the unique national supplier of antiretroviral drugs including lamivudine and tenofovir.

Adherence is the main factor associated with the viral suppression and its maintenance over time; conversely, non-adherence is a risk factor for the emergence of resistant strains [17]. The 15- to 24-year age group was 80% low-adherent. The adherence to a long-term medication is usually a challenge in young subjects. Generally speaking and in relation to the HBV infection, there is no need to rush to initiate antiviral treatment for young people; rather, the emphasis should be put on the clinical follow-up.

The rates of biochemical and virological responses decreased with decreasing adherence [15] [16] [17]. Adherence to treatment would therefore be a predictor of the viral suppression and normalization of transaminases. Consequently, the biochemical and virological responses were better in the context of high adherence.

In our study, 47.6% of the participants under lamivudine molecule developed a resistant strain of the virus by five years of treatment. In the literature, the frequency of the resistance to lamivudine is 70% by five years [12]. In our setting, only a small number of patients could afford to get a laboratory viral load assessment and the proportion of resistance to lamivudine may have been underestimated. No case of resistance to tenofovir has been detected in our study and this finding is confirmed by other studies that did find no or very rare cases of resistance to tenofovir [7] [13].

A significant number of patients participating in the study, lacked laboratory and ultrasound data. Financial reason seemed to be behind this, as the study was designed to analyze routine data. We therefore believe that the description given in this work does not present the whole reality of our cohort. However, what is observed gives us a fairly objective picture of the evolution of a regularly monitored VHB patient.